The coronavirus disease 2019, also known as COVID-19, is causing havoc on the society and economy. As of July 2, 2020, more than 10 million have been infected by COVID-19 and more than 500 thousand have died. As individuals and companies brace for an economic recession, they evaluate their budgets and cut costs in response to this global pandemic. And while financial measures are a good place to start, it is not enough.
Intelligence technologies will help humanity deal with COVID-19. In this blog post, I explain how machine learning and deep learning technologies can be applied to solve humanity’s most pressing challenge.
Disclaimer: I do not claim to know the future. In this blog post, I describe my vision of the future. I describe many possibilities for the future.
COVID-19 Impacts on Society and Economy
To understand how artificial intelligence can be used to deal with COVID-19, it is important to understand the impacts of this coronavirus on society and the economy. The longer that the coronavirus spreads, the impacts start to cascade, whereby one impact causes another. There are three phases of COVID-19: short-term, medium-term, and long-term.
First Phase of Social & Economic Impact
The first phase of COVID-19 is short-term. Global countries are experiencing many of the following:
- Employment insecurity and underemployment. As people’s fear of job loss rises, they change their behavior. They become disengaged at work and their performance suffers. Human resources recognize that job instability is a mental condition and offer mental relief and support. Using machine learning models, modern organizations identify the most vulnerable workforce and employees and offer support accordingly.
- Job loss. As unemployment increases, people are forced to find a new job or re-skill in response to COVID-19. Recruitment companies develop custom machine learning models, powered by recommendation systems, to offer suggestions based on an individual’s job search criteria.
- Increase in poverty risk and working poor. The working poor face risks of poverty amidst COVID-19. Government institutions and organizations use deep learning to find the most vulnerable populations using classification models. Governments develop regression models to better predict COVID-19 outbreaks and respond with financial aid.
- Hunger — food and fuel insecurity. As COVID-19 impacts the food supply chain, food security is threatened. Foodservice and food supply chain companies are forced to adopt in response. Using deep learning, food, and beverage companies better predict demand and adjust supply accordingly.
- Excess mortality and morbidity. COVID-19 kills. The elderly and critically sick are most affected by this pandemic. As the healthcare system struggles to adapt to slow the spread, COVID-19 continues to spread. Healthcare companies and government agencies are in the race against time to save lives. Healthcare companies and health providers develop deep learning algorithms to detect COVID-19 using radiography imagery (x-rays, computed tomography scans, and magnetic resonance imaging tests). Federal and local governments use neural networks and time-series predictions forecast and model the spread of COVID-19.
- Higher infection and death rates of marginalized populations and those with poor health and in territories with fragile health systems. Some populations are at greater risk than others against COVID-19. Using deep learning algorithms, health care companies provide appropriate support. Government agencies and institutions use neural networks to identify understaffed hospitals and issue local aid.
- Increase in gender-based violence. As anxiety rises and people are forced to stay at home under government orders, violence takes to the streets. As people wear coronavirus masks to cover up, they hide their identity and become anonymous. As violent behavior increases, police and enforcement agencies use deep learning to perform facial analysis and facial recognition.
- Increased alcohol consumption. People isolate themselves and keep a social distance. People turn to drugs and alcohol to numb the pain of loneliness. As alcohol consumption rises, companies are forced to adapt. Modern companies offer at-home workout and meal planning services, powered by machine learning models.
- Increase in levels of stress & anxiety. As uncertainty about the future and economy rises, people’s mental states are affected. Rising stress levels and anxiety accelerates mental health problems. Stress builds up from a lack of finances and social instability. Organizations, recognizing these trends, use deep learning technologies to offer personalized mental health support to their employees.
Second Phase of Social & Economic Impact
The second phase of COVID-19 is medium-term. Many countries are experiencing at least one of the following:
- Loss of gender equality gains. As society turns to violence and unrest, gender equality is affected. Companies and organizations turn to machine learning models to find gender inequality by analyzing workforce communication and offer suggestions for managing gender-related issues.
- Rising suicides. Depression is a lonely killer. When depressive thoughts build up, they cause individuals to believe they are not worthy of life. As dark thoughts become the new norm, suicide is around the corner. Modern companies turn to natural language processing to scan workforce communication data and highlight individuals who are more probable to suicide. If the workforce is international, companies use machine translation to understand multiple languages. Text classification, sentiment analysis, and named entity recognition models are developed to understand what’s understand being said, with what tonality and in what context.
- Food shortages. Rising delayed food supply chains and stress levels cause panicked shopping and rising food shortages. Food companies use time series forecasting to estimate future demand and supply of food and beverages. Foodservice companies turn to time series classification models, powered by neural networks, to better understand how the food supply chain is changing over time.
- Increase in avoidable hospitalizations. Many hospitalizations are avoidable if symptoms are detected early. For circumstances where hospitalizations are unavoidable, health care companies are turning to data. Using deep learning, healthcare providers and hospitals can better predict demand, manage staff, and coordinate with appropriate authorities.
- Shortage of informal care and increased isolation of older people. As individuals tighten their finances, they are unable to offer informal care or unpaid care which is provided to older and dependent persons. As social relationships are weakened, spouses, parents, children, other relatives, neighbors, friends, or other non-kin provide for themselves first and foremost. This leaves older and dependent people at risk of lack of funds and a lack of care. Government agencies and personal health companies servicing the elderly develop machine learning models to identify areas of immediate attention and offer financial support.
- Unemployment rises and stays high. As companies lay off employees, this gives rise to higher and higher unemployment. Since fewer companies hire, the unemployment rates stay high. To reduce reliance on employees, companies use deep learning to automate business operations and processes. To build business resilience, organizations develop end-to-end machine learning systems to automate common tasks such as routing emails, routing messages, etc.
- Increase in poverty risk. As people’s basic needs are unmet, they increase their risk of poverty. Government institutions and organizations step it to offer financial support and temporary relief measures. Using machine learning technologies, the government can evaluate millions of people and offer immediate support for most at risk.
- Mental health problems. Rising tensions within families and companies give rise to mental health problems. Managing the health of employees is now possible thanks to machine learning algorithms. Modern organizations use machine learning to monitor the health and wellbeing of their workforce and offer suggestions for improvement.
- Housing insecurity — increasing homelessness. Housing insecurity increases homelessness and induces stress on existing homeowners. Real estate management companies use machine learning to identify the most affected individuals and offer financial support.
- Rising crime. As people revert to crime and vandalism, they are enacting their fears of the future. Police and enforcement agencies better understand crime using video cameras and computer vision algorithms.
- Firm closures. As business models are put to the test, some companies are unable to survive. Some companies are forced to shut down. These firm closures offer an opportunity at market analysis, powered by deep learning technologies.
- Criminal exploitation, loan sharks and recruitment into organized crime. Financial instability causes financial crime. As organized crime is formed, government agencies and tax service institutions identify fraudulent activities using anomaly detection and machine learning algorithms.
- Adverse childhood experiences. Children are affected by COVID-19 since every family is stressed and incurs financial troubles. As organizations lose their employees, children lose their experiences which they once took for granted, such as playing and gathering with other kids.
- Stigma and xenophobia. COVID-19 started in China. The world forms prejudices against people from China. The resurgence of racism and xenophobia is accelerated by COVID-19. Organizations and companies use text classification, relation classification to detect stigma
- Increased family stress. Family stress rises when finances are unpredictable. Family stress rises when families are forced to stay at home. Short of surveilling families, companies can surveil employees. Using computer vision models,
- Disadvantaged children are less able to catch up on schooling. As schools are forced to adopt and implement online schooling, they leave some students behind. Some disadvantaged children are left behind as there is less time for personalized learning. Education institutions respond with recommendation systems able to offer suggestions for which students to offer support based on course progression.
- Rising levels of not in education, employment, or training. As individual confidence drops, the rise of the idle population emerged. Government organizations use deep learning to find these individuals and offer personalized assistance.
- Alcoholism and addition. Rising personal stress gives way to rising additions and the use of alcoholism. Government institutions offer relief to the most isolated populations. Using machine learning algorithms, governments can offer just-in-time support.
Third Phase of Social & Economic Impact
The third phase of COVID-19 is long-term. Some countries are already experiencing at least one of the following:
- Breakdown of social cohesion. As society loses trust for the government, trust for safety and is unable to meet the basic needs of survival (think food, shelter), social cohesion breaks down. Organizations are quick to re-establish trust by giving away free or discounted products or services. Government institutions use machine learning to find areas requiring the most attention.
- Increased inequality. As social inequality rises, organizations are forced to implement countermeasures. Thanks to deep learning and clustering algorithms, organizations can better detect unfair treatments.
- Long-term ill health. As illness cascades into long-term illness, individuals are forced to stay at home and seek medical health. Healthcare companies use computer vision (facial analysis and emotion recognition) and natural language processing (text classification and sentiment analysis) to analyze patients using virtual telemetry.
- Slower recovery and widening economic and health gaps between geographical areas. The economy is slow to recover in response to COVID-19. Financial companies and government agencies use deep learning to perform time series forecasting and predict the future of the economy.
- Increase in avoidable hospitalizations. Fear of contracting others means individuals are forced to go to hospitals when critically ill. Healthcare companies use deep learning to monitor bed-ridden patients and provide the best possible treatments based on real-time symptoms.
- Long-term unemployment. As individuals lose jobs, long-term unemployment becomes the new normal. Companies adapt by using deep learning models to find ideal candidates. Organizations use machine learning to build resilient businesses by finding better ways to use company data.
Artificial intelligence, machine learning, and deep learning offer hope in managing COVID-19. Now is the time to deal with the coronavirus before it does further damage.
At Produvia, we can help! Since 2013, the Produvia team has developed machine learning models to solve the world’s hardest problems.
Interested to solve coronavirus problems?
Tell us about your project and a member of our team will get back to you. Get started!
This post originally appeared on Medium on July 2, 2020.